of circulating antibodies in dogs infected with Dirofilaria immitis. Two groups of ... trols, and clinically diagnosed cases of occult dirofilariasis. Antibody against a.
Vol. 13, No. 2
JOURNAL OF CLINICAL MICROBIOLOGY, Feb. 1981, p. 309-312 0095-1137/81/02-0309/04$02.00/0
Quantitative Fluorescent Immunoassay for Measurement of Antibody to Dirofilaria immitis in Dogs HOWARD J. GITTELMAN'* ROBERT B. GRIEVE,' MARYL M. HITCHINGS,2 RICHARD H. JACOBSON,' AND RAYMOND H. CYPESS' Department of Preventive Medicine, New York State College of Veterinary Medicine, Corneli University, Ithaca, New York 14850,' and International Diagnostic Technology, Santa Clara, California 950502
An automated fluorescent immunoassay technique (FIAX; International Diagnostic Technology, Santa Clara, Calif.) has been developed for the quantitation of circulating antibodies in dogs infected with Dirofilaria immitis. Two groups of sera, group I consisting of 77 samples and group Il consisting of 126 samples, were obtained from experimentally infected microfilaremic dogs, known negative controls, and clinically diagnosed cases of occult dirofilariasis. Antibody against a partially purified trichloroacetic acid-soluble extract of a soluble somatic extract of D. immitis was measured by FIAX and by an enzyme-linked immunosorbent assay (ELISA). A standard curve was drawn from four samples with known FIAX titers and fluorescent signal unit values. The standard curve was used to determine titers of unknown samples. The correlation coefficients determined in the analysis of logio ELISA and logo FIAX values were r = 0.9057 and r = 0.8976 in groups I and II, respectively. Eighty-three percent of the titer values calculated by FIAX in group I were within one dilution, and 95% were within two dilutions, of those titers obtained by ELISA. In group 1, 79 and 96% of the values obtained by FIAX were within one and two dilutions, respectively, of those obtained by ELISA. FIAX proved to be a reproducible and convenient assay for the measurement of serum antibody in dogs experimentally infected with Dirofilaria.
Dirofilaria immitis is a filarial nematode parasite which, as an adult, typically, inhabits the right ventricle, pulmonary artery, and adjacent pulmonary vasculature of dogs. The parasite is widely distributed in warmer climates throughout the world and may cause serious disease and death. The infection in dogs inhabiting endemic areas can be prevented by daily administration of diethylcarbamazine (3, 4). However, the cost or inconvenience of this treatment may be prohibitive in some instances. Patent infections can be treated by killing the adult parasite with thiacetarsamide (2). The toxicity of the arsenical and the consequences due to emboli of dead parasites make thiacetarsamide treatment relatively dangerous. Definitive diagnosis of this infection can be made only in those cases where microfilariae (embryos) of D. immitis are recovered from peripheral blood. In many cases, however, infected dogs have no circulating microfilariae, yet are still subject to damage by the adult parasite. An immune-mediated phenomenon may precipitate the amicrofilaremic state (9); in these instances there is an immune response directed at the microfilarial stage but not at the adult. An indirect fluorescent-antibody test using intact mi-
crofilariae has been used successfully to detect anti-microfilarial antibody in those dogs with this stage-specific immune response (8, 9). Amicrofilaremic infections may result from the presence of only one D. immitis sex in the dog or from an effete infection. Optimally, a serodiagnostic test would detect amicrofilaremic infections regardless of the etiology and would also detect the immature, precardiac stages during which treatment may be instituted without the danger of worn embolism. In this study we were interested in using a partially purified, adult D. immitis-derived antigen in a quantitative indirect fluorescent-antibody test (FIAX; International Diagnostic Technology, Santa Clara, Calif.), for the serodiagnosis of D. immitis infections. This assay was compared with a manual, semiquantitative, enzymelinked immunosorbent assay (ELISA), using the same antigen. MATERLALS AND METHODS Serum samples. Two groups of sera were assayed for anti-D. immitis antibody by FIAX, and the results were compared with those obtained by ELISA, employing the same serosystem. Groups I and II consisted of 77 and 126 samples, respectively. Serum samples were obtained, as described previously (6), from ex309
310
GITTELMAN ET AL.
perimentally infected microfilaremic dogs and their sibling controls, which were maintained in arthropodfree conditions. Sera from dogs with clinically diagnosed cases of occult dirofilariasis were also examined. The samples selected were evenly distributed across ELISA titers of 1:2 through 1:1,024. ELISA procedure. Ail titers were determined by using an ELISA, as described previously (la). Polystyrene Microtiter plates (Dynatech Laboratories, Inc., Alexandria, Va.) with round-bottom wells were used as the support matrix. The antigen was identical to that used in FIAX. Twelve wells were coated with 50 41 of antigen per well at a concentration of 5 ,ug/ml. Samples were serially diluted 12 times with a 50-,ul Microtiter diluter (Dynatech Laboratories, Inc.), and the endpoints were determined by visual inspection. FIAX assay. The support matrix used in the FIAX assay was a key-shaped plastic stick (STIQ; International Diagnostic Technology) with a cellulose acetate-nitrate polymer-coated disk fixed to both sides of the terminal portion. The antigen used in both FIAX and ELISA was a trichloroacetic acid-soluble extract of a soluble somatic extract of adult D. immitis which was purified by gel filtration and cation-exchange chromatography (1). Optimum antigen concentrations were determined in separate experiments by block titration. The antigen was applied at a final concentration of 20 ,ug/ml of 0.1 M carbonate buffer (pH 9.6). Approximately 0.5 gg of antigen was evenly distributed over disk surface no. 1 and allowed to air dry for 2 h. Surface 2 was untreated and used as a nonspecific sample control. Antigen-coated STIQs were stored in a storage cassette at 4'C for up to 1 month. Ail wash steps and dilutions were made with 0.01 M phosphate-buffered saline with 0.15% (vol/vol) Tween 20, pH 7.4. The buffer was refrigerated and either used or discarded within 7 days of preparation. Fluorescein isothiocyanate-conjugated anticanine immunoglobulin G (lot S598, Miles Laboratories, Elkhart, Ind.) was diluted 1:150 with phosphate-buffered saline-Tween buffer. This conjugate concentration was judged to be optimum in separate, standard titration experiments. Four test tube racks, each with four rows of 12- by 75-mm tubes containing the necessary reactants, were placed on a horizontal shaker and maintained at a 450 incline for the duration of the assay. Agitation at a 45° incline ensured uniform application of reagent over the disk surface during the incubation periods and wash steps. Serum samples were diluted 1:41 (15 id of serum plus 600 ,ul of buffer) using the IDT automatic pipettor (International Diagnostic Technology). Samples of known FIAX titer (calibrators) were prepared identically but were run in duplicate. One STIQ per sample was incubated at room temperature in the serum dilution on the shaker for 40 min, with the antigen side down. A 600-jul amount of buffer was dispensed into each tube of racks 2 and 4 for the wash steps, and 500 d1 of diluted conjugate was placed into the tubes of rack 3. Upon completion of the incubation in diluted sera, the STIQs were transferred to the first wash and agitated for 10 min. STIQs were then incubated in conjugate for 30 min at room temperature, followed by a wash in buffer for 10 min. Fluorescence
J. CLIN. MICROBIOL. was measured within 10 min after completion of the final wash. A fluorimeter (International Diagnostic Technology, Santa Clara, Calif.) was used to measure the fluorescence of the STIQ surface. First, the base line of the fluorimeter was adjusted to zero, then the highest titer sample was inserted antigen side up, and the gain was set so that the fluorimeter display read 160. The fluorescent signal units (FSUs) indicated on the display were recorded. The nonspecific fluorescence of each sample was determined by measuring the fluorescence on the side of the STIQ without the antigen. AFSUs were obtained by subtraction of FSUs due to nonspecific fluorescence from FSUs obtained from measurement of the antigen-coated surface. The fluorescence measured from each sample (AFSUs) was proportional to the quantity of anti-D. immitis antibody in that sample. Analysis of data. The data from group I were graphed, with logo AFSU plotted on the ordinate and logo of reciprocal ELISA titer plotted on the abscissa. A line of best fit was determined by least-squares linear regression analysis and was drawn through the plotted values for all positive samples (>1:64); sample titers were calculated by interpolation. Titers were corrected for nonspecific fluorescence by using AFSU values when interpolating from a standard curve or best-fit line. Samples from group I with logo ELISA-logio FIAX coordinates which fell on the line of best fit were chosen as calibrators. Their titers were measured repeatedly to determine variability between assays, the mean titer, and standard deviation. A standard curve was calculated with the data obtained for four of the calibrators, with logo of reciprocal FIAX titer plotted on the abscissa and logo AFSU plotted on the ordinate. Titers of the remaining samples in group I were determined by reference to the standard curve. Titers were determined for the 126 samples in group II, and the data were analyzed to verify the results obtained from group I. Two samples were chosen randomly from group I to determine variability within each assay. Twelve determinations of sample 1 and 11 of sample 2 were made, and their titers were determined in the same assay. The coefficient of variation was calculated for samples 1 and 2, using the data obtained.
RESULTS Classification of samples as negative or positive was based on the distribution of AFSUs illustrated in Fig. 1. Samples with FIAX titers of >1:20 had AFSUs proportional to ELISA titers. However, ail sera with titers of
